So I have been thinking lately, lets say hypothetically I were to want to test out how a couple heatsinks perform against each other. i would not have the same temperature throughout all of my testing so I need a way to correct my results. Lets say I test a cpu at 20 degrees room temperature and test a different heatsink at 25 degrees room temperature. Would i have to simply at +5 degrees to the first temperature to correct the results? Or does it not work that way?